RESEARCH ARTICLE

The Migrant Paradox in Children and the Role of Schools in Reducing Health Disparities: A Cross-Sectional Study of Migrant and Native Children in Beijing, China Ying Ji1, Yanling Wang1, Lei Sun1, Yan Zhang2, Chun Chang1* 1 Department of Social Medicine and Health Education, School of Public Health, Peking University, Beijing, China, 2 Beijing Centers for Disease Control and Prevention, Beijing, China

a11111

* [email protected]

Abstract

Data Availability Statement: All relevant data are within the paper and its Supporting Information files.

Migrants usually exhibit similar or better health outcomes than native-born populations despite facing socioeconomic disadvantages and barriers to healthcare use; this is known as the “migrant paradox.” The migrant paradox among children is highly complex. This study explores whether the migrant paradox exists in the health of internal migrant children in China and the role of schools in reducing children’s health disparities, using a multi-stage stratified cluster sampling method. Participants were 1,641 student and parent pairs from Grades 4, 5, and 6 of eight primary schools in Beijing. The following school types were included: state schools with migrant children comprising over 70% of total children (SMS), private schools with migrant children comprising over 70% (PMS), and state schools with permanent resident children comprising over 70% (SRS). Children were divided into Groups A, B, C or D by the type of school they attended (A and B were drawn from SRSs, C was from SMSs, and D was from PMSs) and whether they were in the migrant population (B, C, and D were, but A was not). Related information was collected through medical examination and questionnaires completed by parents and children. Prevalence of caries, overweight and obesity, poor vision, and self-reported incidence of colds and diarrhea in the previous month were explored as health outcomes. The results partially demonstrated the existence of the migrant paradox and verified the role of schools in lowering health disparities among children; there are theoretical and practical implications for improving the health of migrant children.

Funding: This study was supported by National Social Science Foundation of China (11CGL080). YJ received the funding. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Introduction

OPEN ACCESS Citation: Ji Y, Wang Y, Sun L, Zhang Y, Chang C (2016) The Migrant Paradox in Children and the Role of Schools in Reducing Health Disparities: A CrossSectional Study of Migrant and Native Children in Beijing, China. PLoS ONE 11(7): e0160025. doi:10.1371/journal.pone.0160025 Editor: Koustuv Dalal, Örebro University, SWEDEN Received: December 25, 2015 Accepted: May 27, 2016 Published: July 26, 2016 Copyright: © 2016 Ji et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Competing Interests: The authors have declared that no competing interests exist. Abbreviations: SES, social economic status; SMS, state migrant children primary school; PMS, private

Compared with the native population, migrants often have relatively low social status and income level and insufficient access to health services [1, 2], all of which are health risk factors. However, many health indices, such as self-reported health, pregnancy outcomes, and body weight in infants, are better in migrant populations than in native populations [3–5]. This is

PLOS ONE | DOI:10.1371/journal.pone.0160025 July 26, 2016

1 / 12

Migrant Paradox and the Role of Schools

migrant children primary school; SRS, state permanent resident children primary school.

the “migrant paradox.” This phenomenon and the “healthy migrant effect,” which is used to explain the migrant paradox, have long been the foci of research on the health of international migrants [6]. Similar phenomena have also been observed in internal migration (i.e., migration within a country) [7, 8]. However, some studies in recent years have questioned this interpretation of the data. They have proposed that these observations are linked to the origins of the populations under consideration, as well as demographic characteristics and choices of health indices used in conducting research [9–13]. Children are a special group in migrant populations. On one hand, they live within similar social and family environments as the migrant populations generally, in which low socioeconomic status (SES) generally leads to health disparities between migrant and native children [14–16]. On the other hand, schools can affect migrant children’s health behavior and outcomes [17, 18] and can play an important role in reducing health disparities and improving health equality. Most children spend their time at home and schools. According to Green and Kreuter’s health promotion planning framework [19], environmental contexts might influence children’s health as reinforcing and enhancing factors, such as families and school environments. Schools provide resources to easily intervene in children’s health. Therefore, health disparities in migrant children appear more complex than that of the migrant population in general. However, previous research has mainly examined international migrant children regarding mental health, asthma, obesity, and risky behaviors and addressed the role of family culture, rather than that of schools [5, 20]. In China, migrant children study in various types of schools with various conditions. If the role of schools in children’s health can be demonstrated, settings-based approaches could be used in health interventions in the future. China presents a highly suitable context for examining the internal migrant paradox. With the rapid development of the Chinese economy, the internal migrant population in China has been growing rapidly. China’s “migrant population” is defined as individuals leaving their residences for cities for a certain period (more than six months) without changing household registration. In 2012, the internal migrant population was approximately 236 million in China [21]. Similarly, the number of children in migrant populations is also increasing. According to 2010 data, 20.8% of the migrant population were children under 14 years old [22], whose living conditions were characterized by instability, overcrowding, poor sanitation, disadvantaged schooling, and social and cultural isolation [23]. In 2010 in Beijing, 28% of children 6–14 years of age were migrants. In other words, when providing health services to children 6–14 years of age, one in every four will be a migrant [24]. However, there have been relatively few studies of the health disparities between migrant and permanent resident children; whether migrant paradox exists among school-age children is also unclear. To fill this gap in the literature, the present study explores whether the “migrant paradox” exists in migrant children and analyzes the influence of schools and SES on health disparities among children. Therefore, we assume the phenomena might exist among internal migrant children; that is, migrant children are healthier than resident ones and migrant children in schools with better conditions are healthier than those with poor conditions, controlling for SES. In particular, the analysis of the role of schools may provide a reference for schools to contribute to the reduction of health disparities among children.

Materials and Methods The study was conducted in April and May 2012 in Beijing. A multi-stage stratified cluster sampling method was adopted. In the present study, “migrant children” refers to those with household registrations outside Beijing who had been living in Beijing for over six months. Among the 16 districts in Beijing, the Haidian and Fengtai districts were chosen due to their

PLOS ONE | DOI:10.1371/journal.pone.0160025 July 26, 2016

2 / 12

Migrant Paradox and the Role of Schools

relatively large populations of migrant children, comprising 14.6% and 14.06% of all migrant children in Beijing [24]. In each district, schools with medium student sizes (800–1200) were randomly chosen. One state and one private school with migrant children comprising over 70% of total children in the school was chosen in each district (one state migrant children primary school and one private migrant children primary school, i.e., one SMS and one PMS), paired with two state primary schools of similar sizes in which permanent resident children comprised over 70% of total children in the school (state permanent resident children primary schools, i.e., SRS). In Beijing, state schools (SRSs and SMSs) are government-funded under the unified management of the Board of Education; PMSs are funded by private investment and not under the unified management of the Board of Education. SRSs typically have the best conditions among the three school types, with health education courses regularly offered by qualified teachers and basic hygiene amenities (sufficient taps, washrooms, and classroom lights) and a balanced diet provided. SMSs have poorer facilities and are often located in rural-urban fringe zones. PMSs have relatively poor facilities and management conditions, but are preferred by the migrant population due to their low tuition fees and locations within migrant-populated regions. Two classes from Grades 4, 5, and 6 from each included primary school were selected; all students in these classes, as well as their parents, were included as study participants. Overall, 1,641 pairs of students and parents were obtained from eight primary schools, giving a response rate of 92.3%. Students were divided into four types by their school types and migrant status: Type A represented permanent resident children in SRSs; Type B represented migrant children in SRSs; Type C represented migrant children in SMSs; and Type D represented migrant children in PMSs. Among these, Types B, C, and D all included migrant children, but in varying schooling environments. The variables of sex, residence duration, and urban/rural registered residency were included in the models as basic demographic variables. Residence duration was a factor influencing migrant paradox [12] and the phenomena may gradually disappear with longer residence duration. Urban/rural registered residency was included as migrant origin types as some researchers hypothesized that the migrant paradox observations were linked to the population origins [9,10]. Age was excluded in the model because participants were from 4–6th grades and the age means among the various groups were not significantly different. We selected SES to measure families’ environments; it reflected material circumstances as well as educational environment. In addition, many studies had adopted it as a factor reflecting family conditions [25, 26]. In this study, information on family SES was obtained from questionnaires completed by parents; it included the education levels of the children’s parents, average monthly income per person, ownership of durable consumer goods (television, washing machine, refrigerator, air conditioner, computer, or automobile), accommodation type (apartment block, brick bungalow, or adobe bungalow), and whether bathrooms were shared. Because average monthly income per person is a sensitive question, its non-response rate reached 20%. Consequently, in this study, ownership of durable consumer goods was used to replace this variable. Previous research has applied this method [27, 28]. Factor analysis was used in this study, including parents’ education level (no school education, primary school, junior high school or technical middle school, senior high school or technical school, junior college, and university and above were given 1–6 points respectively; the parent with the higher education level was recorded); number of durable consumer goods (television, washing machine, refrigerator, air conditioner, computer, or automobile; the number owned was equal to the mark given, with a maximum of 6 points); and living conditions (consisting of the two variables of accommodation type and whether bathrooms were shared; accommodation type was divided into apartment block, brick bungalow, and adobe bungalow, which were given 3, 2, or 1 points respectively. In China, apartment blocks typically provide better accommodation

PLOS ONE | DOI:10.1371/journal.pone.0160025 July 26, 2016

3 / 12

Migrant Paradox and the Role of Schools

conditions and reflect better economic status than brick bungalows. In Beijing, a small number of people of very low economic status live in adobe bungalows). Bathrooms being shared or not shared was given 1 or 2 points respectively; this resulted in a maximum of 5 points for living conditions). Variables of these three factors underwent factor analysis to produce SES scores. The construct validity of extraction factors was a = 0.62; the Cronbach’s alpha of the questionnaires was 0.81. The equation was: SES = (0.428 × parents’ education level score + 0.429 × number of durable consumer goods + 0.414 × living conditions score). Factor analysis then transformed SES scores into a standard normal distribution with mean = 0 and standard deviation = 1. Higher scores indicated higher SES. We selected and measured health indices as follows. Parents reported whether or not children had experienced colds or diarrhea in the past month. Doctors, who received standardized training, conducted medical examinations of the children, including height, weight, caries, and vision. The examination method was based on Technical Standard for Physical Examination for Students, published by the People's Republic of China Ministry of Health and the Standardization Administration of China [29]. The five health indices were selected based on common diseases among schoolchildren in China. Most colds and diarrhea were acute communicable diseases; obesity, poor vision, and caries were common chronic diseases regularly checked in physical examinations in state schools in China. Height was measured as standing body height with 0.1 cm accuracy using a height meter (Model: 2m; Beijing Dong Hua Teng Sports Equipment Co., Ltd). Weight was measured using a leveraged body weight scale with 0.1 kg accuracy (Model: RGT-140; Wuxi Weighing Apparatus Manufacturer).Overweight and obesity were determined by age and sex based on Body Mass Index reference norm for screening overweight and obesity in Chinese children and adolescents by the Working Group on Obesity of China [30]. By this standard, a BMI of 18.1 is considered as the threshold of overweight and obesity in 8-year-old children. The thresholds for 9-year-old boys and girls are 18.9 and 19.0 respectively; those for 10-year-old boys and girls are 19.6 and 20.0 respectively; those for 11-year-old boys and girls are 20.3 and 21.1 respectively; those for 12-year-old boys and girls are 21.0 and 21.9 respectively; those for 13-year-old boys and girls are 21.9 and 22.6 respectively. Caries in this study referred to those in both deciduous and permanent teeth. Observation of any one of three types of cases (current caries, filled caries, and lost teeth due to caries) was considered observation of caries. Standard logarithmic distance vision tables were used for vision examination, with naked eye vision < 5.0 defined as “poor vision.” Caries and vision were examined by trained dentists and ophthalmologists. Each health index underwent classification calculation by the four groups of children. Continuous variables were described using means and standard deviations; analyses of variance were used to compare groups. Categorical variables were described using percentages and chisquare tests were used to compare groups. Each health outcome was analyzed to examine the effect of SES and school type using logistic regression with two models. Model 1 included sex, registered residency, residence duration, and family SES; Model 2 was based on Model 1, but added groupings of children. Group B was of particular interest: Group B children were in the same school type as Group A children, but were in the same household type as Groups C and D children. Group B was therefore used as the control group in the regression analysis of dummy variables. Odds ratios (ORs) and changes in the 95% CI of SES in these two models were used to determine the effects of schools and children’s identities on health indices. Cox and Snell’s R square and the Hosmer-Lemeshow Goodness of Fit index were used to assess model fit. All analyses were conducted using SPSS 13.0 (SPSS Inc., Chicago, IL). The Institutional Review Board at Peking University approved this study (IRB0000105212012). Written informed consent was obtained from the participating students and their parents in this study prior to administration of the survey.

PLOS ONE | DOI:10.1371/journal.pone.0160025 July 26, 2016

4 / 12

Migrant Paradox and the Role of Schools

Table 1. Basic demographic characteristics of study participants. χ2 / F

P

184 (56.1)

24.97

The Migrant Paradox in Children and the Role of Schools in Reducing Health Disparities: A Cross-Sectional Study of Migrant and Native Children in Beijing, China.

Migrants usually exhibit similar or better health outcomes than native-born populations despite facing socioeconomic disadvantages and barriers to hea...
223KB Sizes 1 Downloads 7 Views